702 RICE ART. L 



same right up to the surface as throughout the solution. We 

 will denote this excess per unit area by Ti. There will of course 

 be other ions present; positive ions such as those of other metals 

 and of hydrogen; negative ions such as sulphion, chlorion, etc. 

 The total charge on all these ions, positive and negative, must 

 be equal and opposite to the charge on the mercury side of the 

 surface, so that if there is a deficiency of electrons in the mercury 

 the negative ions must preponderate in the solution part 

 of the discontinuous region; i.e., 



Te/?, + TiiSi + r2i82 + . . . = 0, 



where 2, .... refer to ions other than the mercury ions. We 

 now have the equation 



d<T = - rjadt - Te dM^ - Ti dMi - Tg dMi - . . • , 



or, at constant temperature, 



da = - ^,V, dV - Ti dMi - T2 dMi - ... 



= - qdV - Ti dMi - Ta dikfg - • • . 



This formula exhibits the Lippmann term —q dV(q is the charge 

 per unit area on the mercury) and Gibbs terms in addition for 

 the various ions present in excess or deficiency on the solution 

 side of the dividing surface. These are the specifically adsorbed 

 ions, cations or anions, whose influence causes the deviations 

 from the simple normal state of affairs covered by the Lippmann 

 term alone. Thus the simple criterion that at the condition 

 for maximum a the charge should be zero is not necessarily 

 true, since for that condition it is the expression 



dMi dMi 



which is zero. If we assume that 8M1, 8M2, etc., are all altered 

 by ^idV, jSa^F, etc., respectively, we would, on account of the 

 fact that Sr/3 = 0, obtain the result that da is always zero, 

 which is absurd. Or we might assume that some of the MrS 

 alter by ^rSV (say the Mi for the mercury ion because it is a 



